The ISME Journal (2013) 7, 1661–1664 & 2013 International Society for Microbial Ecology All rights reserved 1751-7362/13 www.nature.com/ismej SHORT COMMUNICATION Diverse populations of lake water exhibit chemotaxis towards inorganic nutrients

Paul G Dennis1,2, Justin Seymour3, Kimber Kumbun2 and Gene W Tyson1,2 1Australian Centre for Ecogenomics, School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, Queensland, Australia; 2Advanced Water Management Centre, The University of Queensland, Brisbane, Queensland, Australia and 3Plant Functional Biology and Climate Change Cluster, University of Technology, Sydney, New South Wales, Australia

Chemotaxis allows microorganisms to rapidly respond to different environmental stimuli; however, understanding of this process is limited by conventional assays, which typically focus on the response of single axenic cultures to given compounds. In this study, we used a modified capillary assay coupled with flow cytometry and 16S rRNA gene amplicon pyrosequencing to enumerate and identify populations within a lake water microbial community that exhibited chemotaxis towards ammonium, nitrate and phosphate. All compounds elicited chemotactic responses from populations within the lake water, with members of exhibiting the strongest responses to nitrate and phosphate, and representatives of the Variovorax, Actinobacteria ACK-M1 and Methylophilaceae exhibiting the strongest responses to ammonium. Our results suggest that chemotaxis towards inorganic substrates may influence the rates of biogeochemical processes. The ISME Journal (2013) 7, 1661–1664; doi:10.1038/ismej.2013.47; published online 21 March 2013 Subject Category: Microbial population and community ecology Keywords: chemotaxis; microbial communities; 16S rRNA gene; inorganic substrates

In aquatic environments, organic and inorganic sulphur compounds (Miller et al., 2004). In contrast, nutrients are distributed heterogeneously, with chemotaxis towards inorganic substrates is poorly microscale hotspots found throughout the water characterised despite bacterial growth in aquatic column (Azam, 1998; Blackburn et al., 1998). Micro- ecosystems often being limited by dissolved inor- bial cells are generally unable to uptake nutrients ganic nutrients (Church, 2008), and the requirement that are more than a few cell diameters away; for some inorganic chemicals in anaerobic respira- therefore, motility is a key determinant of the size tion (Thauer et al., 1977). In addition, as chemotaxis of the microhabitats within which different popula- studies have focussed on single populations of tions perceive and exploit nutrient sources (Stocker, cultivated bacteria, the generality of this trait is 2012). Non-motile cells can explore B80 nl water unknown. To date, chemotaxis towards inorganic per day, which when compared with ca. 1 ml per substrates has been observed towards: (1) ammo- day for motile cells is very small (Stocker, 2012). nium and nitrate by the marine cyanobacterium, Furthermore, as motility is typically associated with Synechococcus (Willey and Waterbury, 1989); (2) the ability of cells to sense and direct movement ammonium by the purple non-sulphur photosyn- along chemical gradients (chemotaxis), many motile thetic bacterium Rhodobacter sphaeroides (Ingham populations are not only able to explore large and Armitage, 1987; Poole and Armitage, 1989); (3) volumes of water, but are also able to move directly nitrate and nitrite by Shewanella putrefaciens towards perceived resource hotspots (Fenchel, 2002; (Nealson et al., 1995) and several strains of deni- Stocker et al., 2008; Stocker and Seymour, 2012). trifying bacteria (Kennedy and Lawless, 1985; Lee Aquatic bacterial chemotaxis occurs in response et al., 2002) and (4) phosphate by a heterotrophic to a wide range of organic substrates, including marine Thalassospira sp. (Hu¨ tz et al., 2011). amino acids (Barbara and Mitchell, 2003), sugars In this study, we used a syringe-based assay (Malmcrona-Friberg et al., 1990) and organic (Supplementary Figure S1), inspired by the tradi- tional capillary assay (Adler and Dahl, 1967), coupled with flow cytometry and 16S rRNA gene amplicon Correspondence: G Tyson, Australian Centre for Ecogenomics and pyrosequencing to enumerate and identify popula- Advanced Water Management Centre, School of Chemistry and tions within a lake water microbial community that Molecular Biosciences, The University of Queensland, Brisbane, exhibited chemotaxis towards ammonium, nitrate Queensland 4072, Australia. E-mail: [email protected] and phosphate. Chemotaxis assays were performed Received 16 October 2012; revised 11 February 2013; accepted 12 by submerging 1-ml syringes containing 80 mlof0.1M February 2013; published online 21 March 2013 chemoattractant (nitrate/ammonium/phosphate) into Bacterial chemotaxis towards inorganic nutrients PG Dennis et al 1662 lake water for 30 min (Supplementary information). Each chemoattractant was prepared in 0.2 mm- Circle size equals filtered lake water to ensure that the chemical cell count 0 2.5 k Ammonium Nitrate Control Lake water characteristics of the background solution in the Phosphate ; Sphingobacteriales 1 syringe was similar to that in the lake. Filtered lake Sphingobacteria; water control assays with no added chemoattractant Sphingobacteriales Sphingobacteriales 2 were performed in parallel and facilitated the Sphingobacteriales 3 measurement of stochastic movement of cells into our assays (Supplementary Figure S2). All assays ; were replicated six times, providing three samples Variovorax for cell counting and three samples that were pooled Burkholderiales for 16S rRNA gene amplicon sequencing. This novel Polynucleobacter approach facilitated culture-independent character- Methylophilales Methylophilaceae isation of chemotactic populations within a lake Gammaproteobacteria Sinobacteraceae water community. To ensure that counts in the Enterobacteriaceae assays reflected chemotaxis rather than growth OP3 OP3 stimulation, we enumerated cells in 0.2 mm-filtered Verrucomicrobia Verrucomicrobiales Actinobacteria; ACK-M1 1 lake water with and without added nutrients. The Actinomycetales controls were incubated for 30 min, but the tips of ACK-M1 2 the syringes were not submerged in lake water Cyanobacteria Dolichospermum Cyanobacteria 1 (Supplementary Figure S3). Cyanobacteria 2 Cell counts in the nutrient-spiked chemotaxis assays were approximately six times higher than in the control assays (Po0.001, generalized linear Figure 1 Bacterial operational taxonomic units (OTUs) present model; Supplementary Figures S2 and S3), but did at 41% abundance in any sample. The size of each circle not differ from one another. Cell counts in nutrient represents the relative abundance of each OTU normalised by the stimulation assays, however, did not differ from total cell count within each sample. The OTUs are organised by those associated with the filtered lake water control phylogenetic relatedness based on a maximum likelihood boot- (Supplementary Figure S3). These results indicate strapped tree (1000 iterations) of full-length sequences that were identified as the nearest BLAST matches for each OTU. Bootstrap that nitrate, ammonium and phosphate elicited values are represented as follows: 75–100% (closed circles), 50– strong chemotactic responses from at least one 74% (open circles), o50% (no circle). The compositional population within the lake water community and similarity of each community is represented at the base of the that slight differences in the diffusion coefficients of heatmap by a complete-linkage cluster analysis tree of OTU the attractants (ca. 0.17, 0.18 and 0.05 Â 104 cm2 s À 1 abundances. À þ 3 À for NO3 ,NH4 and PO4 , respectively) did not lead to differences in the strength of response. Three members of the Sphingobacteriales exhib- 0.6 ited the strongest chemotactic response to nitrate Control and phosphate (Figures 1 and 2). Phylogenetic 0.4 analysis (Supplementary information) revealed that these Sphingobacteriales populations belonged to Nitrate the family Chitophagaceae. Within this family, 0.2 members of the genera Niastella and Chitinophaga Sphingobacteriales 3 Sphingobacteriales 1 are known to exhibit gliding motility (Lim et al., 0 Phosphate 2009; Weon et al., 2009; Del Rio et al., 2010), and Methylophilaceae ACK-M1 1 Sphingobacteriales 2 some Chitinophaga species are known to reduce PC2 (11.4%) -0.2 nitrate as a terminal electron acceptor (Lim et al., Variovorax 2009). Interestingly, the only other Sphingobacter- iales population observed in the lake water at 41% -0.4 relative abundance, a Pedobacter-like organism, did not exhibit chemotaxis towards nitrate (Figure 1). -0.6 Ammonium This is consistent with observations that Pedobacter isolates do not reduce nitrate (Steyn et al., 1998). Therefore, the positive chemotactic response -0.50 0.5 towards nitrate observed for the Sphingobacteriales PC1 (88.0%) populations may reflect their ability to detect and Figure 2 PCA ordination highlighting differences in operational exploit nitrate as an alternative electron acceptor. taxonomic units (OTU) abundance between the chemotaxis Chemoattraction may also reflect the ability for cells assays. OTU abundances were calculated by normalising the relative abundances of each OTU by the total cell count within to move towards nitrate in response to inorganic each sample. The OTUs are represented as crosses, with the nutrient deficiency; however, there are examples, taxonomic affiliation of the most discriminating OTUs high- where bacteria do not metabolise attractants. lighted in text.

The ISME Journal Bacterial chemotaxis towards inorganic nutrients PG Dennis et al 1663 For example, Escherichia coli is attracted to, but changes in concentrations of chemical attractants does not metabolise methyl-aspartate (Adler, 1969). or repellents is based on E. coli and Salmonella Our results indicate that chemotaxis towards typhimurium (Falke and Hazelbauer, 2001). Current inorganic phosphate is exhibited by a range of evidence indicates, however, that chemosensory bacteria, in particular, the Sphingobacteriales, systems that mediate microbial chemotaxis differ which also showed high levels of chemotaxis considerably between species (Montrone et al., 1998; towards nitrate (Figures 1 and 2). Previously, Boin et al., 2004; Szurmant and Ordal, 2004; Meier chemoattraction to phosphate has only been demon- et al., 2007). The extent of these knowledge gaps strated in a free-living planktonic bacterium belong- highlights that our understanding of chemotaxis in ing to the genus Thalassopira (Hu¨ tz et al., 2011), natural environments is far from complete. which was isolated from the oligotrophic eastern Our study indicates that a wide range of lake Mediterranean Sea. Interestingly, no differences in water bacterial populations have the capacity to the swimming speed of the Thalassopira species move towards inorganic nutrients. This observation was observed when exposed to organic carbon or was previously overlooked due to culture-based inorganic phosphate (Hu¨ tz et al., 2011), suggesting analyses. Chemotaxis towards inorganic nutrients that its motility was not differentially regulated by is a behaviour that likely enables them to exploit the type of nutrient supplied. This observation is microscale gradients of limiting resources in the consistent with our data, which showed that the environment. It has previously been suggested populations responding to nitrate were similar to that bacterial chemotactic exploitation of micro- those that respond to phosphate (Figures 1 and 2). scale patches of organic compounds in marine This may be indicative of cross-regulation of habitats can influence carbon cycling processes and chemotaxis machinery by nitrate and phosphate influence microbial community interactions sensory systems or environmental co-dependency (Azam, 1998; Blackburn et al., 1998; Fenchel, 2002). on these compounds. Our results substantially Our observations indicate that strong, but phylogen- expand the known diversity of bacteria that exhibit etically variable, chemotactic responses towards chemotaxis towards phosphate. inorganic compounds may similarly influence nutri- Ammonium elicited the strongest response among ent cycling processes and microbial competitive the tested inorganic nutrients for the majority of interactions within aquatic ecosystems. dominant populations, although none of these responses were as strong as those observed for nitrate and phosphate by the Sphingobacteriales (Figure 1). Conflict of Interest Organisms closely related to Variovorax, Actinobac- The authors declare no conflict of interest. teria ACK-M1, also known as the acI group (Warnecke et al., 2004), and a member of Methylophilaceae exhibited the strongest responses to ammonium Acknowledgements (Figures 1 and 2). has been We thank Philip Hugenholtz for advice on the manu- previously reported to swarm in response to ammo- script. PGD acknowledges financial support from the nium (Jamieson et al., 2009), and ammonium is Commonwealth Scientific and Industrial Research Orga- known to influence the composition of Actinobac- nisation. GWT is supported by an ARC Queen Elizabeth II teria ACK-M1 assemblages in lakes (Newton et al., fellowship (DP1093175). 2007); however, we are not aware of studies demon- strating that representatives of the Methylophilaceae and Actinobacteria ACK-M1 group exhibit chemo- taxis towards ammonium. This lack of information highlights that our approach has the potential to References greatly improve knowledge of the microbial lineages Adler J. (1969). Chemoreceptors in bacteria. Science 166: that exhibit chemotaxis towards a wide range of 1588–1597. compounds in natural environments. Our assay Adler J, Dahl MM. (1967). A method for measuring the should also facilitate functional characterisation of motility of bacteria and for comparing random and assemblages of chemotactic populations when non-random motility. J Gen Microbiol 46: 161–173. coupled with metagenomics and metatranscrip- Azam F. (1998). Microbial control of oceanic carbon flux: tomics. This approach would help to reveal the range the plot thickens. Science 280: 694–696. and prevalence of mechanisms of movement and Barbara GM, Mitchell JG. (2003). Marine bacterial organi- environmental sensing used by chemotactic popula- sation around point-like sources of amino acids. FEMS tions in natural environments. Gliding motility, for Microbiol Ecol 43: 99–109. example, is used by Myxococcus xanthus during Blackburn N, Fenchel T, Mitchell JG. (1998). Microscale nutrient patches in plankton habitats shown by chemotaxis to phosphatidylethanolamine (Kearns chemotactic bacteria. Science 282: 2254–2256. and Shimkets, 1998), but it is unknown whether this Boin MA, Austin MJ, Hase CC. (2004). Chemotaxis in type of motility is common among chemotactic Vibrio cholerae. FEMS Microbiol Lett 239: 1–8. microorganisms. Similarly, much of what is known Caporaso JG, Bittinger K, Bushman FD, DeSantis TZ, about how microbial chemoreceptors respond to Andersen GL, Knight R. (2010). PyNAST: a flexible

The ISME Journal Bacterial chemotaxis towards inorganic nutrients PG Dennis et al 1664 tool for aligning sequences to a template alignment. toward dinoflagellate products. Appl Environ Micro- Bioinformatics 26: 266–267. biol 70: 4662–4701. Church MJ. (2008). Resource control of bacterial dynamics Montrone M, Eisenbach M, Oesterhelt D, Marwan W. in the sea. In: Kirchman D (ed). Microbial Ecology of (1998). Regulation of switching frequency and bias of the Ocean. 2nd edition, John Wiley & Sons, Inc: the bacterial flagellar motor by CheY and fumarate. Hoboken, NJ, USA. J Bacteriol 180: 3375–3380. Del Rio TG, Abt B, Spring S, Lapidus A, Nolan M, Tice H Nealson KH, Moser DP, Saffarini DA. (1995). Anaerobic et al. (2010). Complete genome sequence of electron acceptor chemotaxis in Shewanella putrefa- Chitinophaga pinensis type strain (UQM 2034T). ciens. Appl Environ Microbiol 61: 1551–1554. Stand Genomic Sci 2: 87–95. Newton RJ, Jones SE, Helmus MR, McMahon KD. (2007). Falke JJ, Hazelbauer GL. (2001). Transmembrane signaling Phylogenetic ecology of the freshwater Actino- in bacterial chemoreceptors. Trends Biochem Sci 26: bacteria acI lineage. Appl Environ Microbiol 73: 257–265. 7169–7176. Fenchel T. (2002). Microbial behaviour in a heterogenous Poole PS, Armitage JP. (1989). Role of metabolism in the world. Science 296: 1068–1071. chemotactic response of Rhodobacter sphaeroides to Hu¨ tz A, Schubert K, Overmann J. (2011). Thalassospira ammonia. J Bacteriol 171: 2900–2902. sp. isolated from the oligotrophic eastern Mediterra- Steyn PL, Segers P, Vancanneyt M, Sand P, Kersters K, nean Sea exhibits chemotaxis toward inorganic phos- Joubert JJ. (1998). Classification of heparinolytic phate during starvation. Appl Environ Microbiol 77: bacteria into a new genus, Pedobacter, comprising 4412–4421. four species: Pedobacter heparinus comb. nov., Ingham CJ, Armitage JP. (1987). Involvement of transport Pedobacter piscium comb. nov., Pedobacter africanus in Rhodobacter sphaeroides. J Bacteriol 169: sp. nov. and Pedobacter saltans sp. nov. Proposal of 5801–5807. the family fam. nov. Int J Syst Jamieson WD, Pehl MJ, Gregory GA, Orwin PM. (2009). Bacteriol 48: 165–177. Coordinated surface activities in Variovorax Stocker R. (2012). Marine microbes see a sea of gradients. paradoxus EPS. BMC Microbiol 9: 124. Science 338: 628–633. Kearns DB, Shimkets LJ. (1998). Chemotaxis in a gliding Stocker R, Seymour JR. (2012). Ecology and physics of bacterium. Proc Natl Acad Sci USA 95: 11957–11962. bacterial chemotaxis in the ocean. Microbiol Mol Biol Kennedy MJ, Lawless JG. (1985). Role of chemotaxis in the Rev 76: 792–812. ecology of denitrifiers. Appl Environ Microbiol 49: Stocker R, Seymour JR, Samadan A, Hunt DH, Polz M. 109–114. (2008). Rapid chemotactic response enables marine Lee DY, Ramos A, Macomber L, Shapleigh JP. (2002). Taxis bacteria to exploit ephemeral microscale nutrient response of various denitrifying bacteria to nitrate and patches. Proc Natl Acad Sci USA 105: 4209–4214. nitrite. Appl Environ Microbiol 68: 2140–2147. Szurmant H, Ordal GW. (2004). Diversity in chemotaxis Lim JH, Baek SH, Lee ST. (2009). Ferruginibacter alkali- mechanisms among the bacteria and archaea. Micro- lentus gen. nov., sp. nov. and Ferruginibacter lapsina- biol Mol Biol Rev 68: 301–319. nis sp. nov., novel members of the family Thauer RK, Jungerman K, Decker K. (1977). Energy ‘Chitinophagaceae’ in the phylum Bacteroidetes, iso- conservation in chemotrophic anaerobic bacteria. lated from freshwater sediment. Int J Syst Evol Bacteriol Rev 41: 100–180. Microbiol 59: 2394–2399. Warnecke F, Amann R, Pernthaler J. (2004). Actinobacter- Malmcrona-Friberg K, Goodman A, Kjelleberg S. (1990). ial 16S rRNA genes from freshwater habitats cluster in Chemotactic responses of marine Vibrio sp. Strain S14 four distinct lineages. Environ Microbiol 6: 242–253. (CCUG 15956) to low-molecular weight substances Weon HY, Yoo SH, Kim BY, Son JA, Kim YJ, Kwon SW. under starvation and recovery conditions. Appl (2009). Niabella ginsengisoli sp. nov., isolated from Environ Microbiol 56: 3699–3704. soil cultivated with Korean ginseng. Int J Syst Evol Meier VM, Muschler P, Scharf BE. (2007). Functional Microbiol 59: 1282–1285. analysis of nine putative chemoreceptor proteins in Willey JM, Waterbury JB. (1989). Chemotaxis toward Sinorhizobium meliloti. J Bacteriol 189: 1816–1826. nitrogenous compounds by swimming strains of Miller TR, Hnilicka K, Dziedzic A, Desplats P, Belas R. marine Synechococcus spp. Appl Environ Microbiol (2004). Chemotaxis of Silicibacter sp. strain TM1040 55: 1888–1894.

Supplementary Information accompanies this paper on The ISME Journal website (http://www.nature.com/ismej)

The ISME Journal